Sodium borohydride-carboxylic acid systems. Useful reagents for the

J. Org. Chem. , 1975, 40 (23), pp 3453–3456. DOI: 10.1021/ .... Dibyendu De, Frances M. Krogstad, Larry D. Byers, and Donald J. Krogstad. Journal of...
9 downloads 0 Views 584KB Size
J.Org. Chem., Vol. 40, No. 23, 1975 3453

Notes Our results demonstrate that hydrogen bromide addition to l a and l b is regiospecific as well as stereospecific, whereas the addition of IC is nonregiospecific. T h e most convincing rationale of our results is depicted in Scheme I. We suggest that l a or l b reacts with hydrogen bromide in a reversible process to form the corresponding unstable aproton complex (2a or 2b), which readily breaks up to form a planar resonance stabilized a-allylic cation (3a or 3b). The addition of bromide anion to 3a or 3b can give rise to the observed allylic bromide (4a or 4b) in each case. The stereospecificity observed can be explained as the cis configuration of 3a or 3b is much more stable than its trans configuration in a nine- or ten-membered sing. In the case of IC we argue that the a-proton complex (2c) is stable, and it is attacked by the nucleophile (Br-) both a t central and terminal centers to form 1-bromocyclotridecene (5) and 3bromocyclotridecene (6), respectively. However, the possibility of the formation of the observed products (5 and 6) via nonplanar allylic and vinylic cations cannot be ruled out completely. In conclusion, the reactions reported here represent the first example of change of orientation of hydrogen bromide addition with change in the ring size. The mode of addition of hydrogen bromide to strain-free 1,2-~yclotridecadiene (IC) resembles that of simple 1,3-disubstituted acyclic alle ne^.^ We propose that the strain factor could be responsible for the observed difference in behavior of C-9 or C-10 as compared to C -13 allene in hydrogen bromide addition.

Experimental Section All boiling points are uncorrected. Ir spectra were recorded on a Perkin-Elmer IR-137 using neat liquids. NMR spectra were recorded on a Varian Model A-60 NMR spectrometer relative to internal standard Me&. The mass spectral measurements were performed by the Mass Spectrometry Laboratory, National Chemical Laboratory, Pooma, India. The gas-liquid chromatography utilized a Varian Model 90-P gas chromatograph with a thermal conductivity detector. Elemental analyses were performed by the Microanalysis Laboratory, Department of Chemistry, Indian Institute of Technology, Kanpur 208016, India. General Procedure for the Addition of Hydrogen Bromide to Cyclic Allenes. The cyclic allene (0.05 mol) was taken in a three-necked round-bottomed flask and cooled to around 15-20' in a nitrogen atmosphere. Hydrogen bromide solution in acetic acid (40% w/v, 12.0 ml, 0.055 mol) was added dropwise with magnetic stirring over a period of 30 min. After the addition was over, it was allowed to stir for another 2 hr. The reaction mixture was poured into 200 ml of water, neutralized carefully with sodium bicarbonate, and extracted with petroleum ether (bp 40-60O). The combined extract was washed thoroughly with water and dried over anhydrous MgS04. Removal of solvent and distillation under vacuum gave the monobromo adduct. Addition of Hydrogen Bromide to 12-Cyclononadiene. From 1,2-cyclononadiene (6.1 g, 0.05 mol) and hydrogen bromide (0.055 mol), there was obtained 8.8 g (86%)of cis-3-bromocyclononene: bp 87-88O (5 mm) [lit.5bp 34-35' (0.05 mm)]; ir (neat) 2018, 1635, and 710 cm-l; NMR (CDC13) 6 5.60 (m, 2 H), 5.00 (br m, 1 H), and 1.00-2.40 (m, 12 H). Anal. Calcd for CgH16Br: C, 53.20; H, 7.39. Found: C, 53.41; H, 7.21. Addition of Deuterium Bromide to 1,Z-Cyclononadiene. Following the general procedure, the treatment of 1,2-cyclononadiene (1.22 g, 0.01 mol) with deuterium bromide (0.012 mol) in acetic acid-d provided 1.5 g (74%) of 3-bromocyclononene-2-d: bp 90' (5 mm); ir (neat) 2020, 1636, and 712 cm-'; NMR (CDC13) 6 5.60 (m, 1 H),5.00 (br m, 1 H),and 1.00-2.40 (m, 1 2 H); mass spectrum m/e 203 and 205 (M+)of almost equal intensity. Addition of Hydrogen Bromide to 1,Z-Cyclodecadiene. Treatment of 1,2-cyclodecadiene (6.8 g, 0.05 mol) with hydrogen bromide (0.055 mol) in acetic acid gave 8.0 g (75%)of cis-3-bromocyclodecene: bp 91-92' (5 mm) [lit.5 bp 86-87O (3 mm)]; ir (neat) 2018, 1634, and 712 cm-'; NMR (CDC13) 6 5.58 (m, 2 H), 5.00 (br m, 1H), and 1.00-2.42 (m, 14 H). Anal. Calcd for ClOH17Br: C, 55.30; H, 7.83. Found: C, 55.13; H, 7.71.

Addition of Hydrogen Bromide to 1,Z-Cyclotridecadiene. 1,2-Cyclotridecadiene (9.0 g, 0.05 mol) was treated with hydrogen bromide (0.055 mol) to yield 9.0 g (70%)of a mixture of 3-bromocyclotridecene and 1-bromocyclotridecene, bp 93-101' (1mm). Careful GLC analysis (10% silicone rubber SE-30, 5 ft X 0.25 in., looo, 30 ml/min Nz) of the reaction product showed two closely situated peaks having 1-bromocyclotridecene (shorter retention time) and 3-bromocyclotridecene in the ratio 45:55. The two components were separated by GLC and compared with authentic sample^.^.^ An authentic sample of 3-bromocyclotridecene was prepared from cyclotridecene (cis and trans mixture) and N-bromosuccinimide:5 ir (neat) 2014, 1636, 970, and 710 cm-'; NMR (CDC13) 6 5.56 (m, 2 H), 4.96 (br m, 1 H), and 0.96-2.45 (m, 20 H). An authentic sample of 1-bromocyclotridecene was made by lithium aluminum hydride reduction of 2,3-dibromocyclotridecene:6 ir (neat) 2016, 1636, 850, and 820 cm-l; NMR (CDC13) 6 5.50 (t, J = 7.0 Hz, 1 H)and 0.98-2.54 (m, 22 H). Anal. Calcd for C13H23Br: C, 60.23; H, 8.88. Found: C, 59.98; H, 8.50. Registry No.-la, 1123-11-1; lb, 4415-98-9; IC, 5601-67-2; 4a, 33332-75-1;4b, 56412-17-0;5, 56412-18-1;6, 38916-95-9; hydrogen bromide, 10035-10-6;deuterium bromide, 13536-59-9; 3-bromocyclononene-2-d, 56412-19-2; cis-cyclotridecene, 2484-66-4; transcyclotridecene, 2484-65-3; 2,3-dibromocyclotridecene, 34833-29-9; N-bromosuccinimide, 128-08-5.

References and Notes (1) Visiting Professor, Department of Organic Chemistry, Indian Institute of

Science, Bangalore 560012,India, 1974-1975.Address correspondence to author at the Indian Institute of Science.

(2)(a) M. C. Caserio in "Selective Organic Transformations", Vol. 1, B. S. Thyagarajan, Ed., Wiley-lnterscience, New York, N.Y., 1970;(b) W. R. Moore, H. W. Anderson, and S. D. Clark, J. Am. Chem. Soc., 95, 835 (1973);(c)T. L. Jacobs and R. C. Kammerer, ibid., 96, 6213 (1974);(d) R. D.Bach and H. W. Holubka, ibid., 96,7614 (1974). (3)R. K. Sharma, B. A. Shoulder, and P. D. Gardner, J. Org. Chem., 32, 241 (1967). (4)G. Nagendrappa, P. Mohanakrishnan,G. Mehta, and D. Devaprabhakara, Chem. Ind. (London), 850 11971). (5) R. W. Fawceti and J. 0. Harris, J. Chem. SOC.,2673 (1954). (6)J. Casanova and B. Wagell, Bull SOC.Chim. Fr.. 2669 (1972). (7)A. V. Fedorova, J. Gen. Chem. USSR (Engl. Trans/.). 33,3508 (1963).

Sodium Borohydride-Carboxylic Acid Systems. Useful Reagents for the Alkylation of Amines Paolo Marchini, Gaetano Liso, and Antonia Reho Istituto di Chimica Organica, Facoltd di Farmacia, Universitb di Bari, Bari, Italy Felice Liberatore* and Franco Micheletti Moracci Cattedra di Chimica Farmaceutica Applicata, Universitb di Roma, Roma, Italy Received March 25, 1975 The recent paper by Gribble et a1.l concerning the alkylation of aromatic amines with liquid carboxylic acids and sodium tetrahydroborate has prompted us to publish our own data on the same reaction, since they not only support the findings of the authors cited, but extend the scope of the reaction to the N-alkylation of amines with solid carboxylic acids, and provide further insight on the possible reaction pathway. In the course of an investigation aimed a t determining the reactivity of some l,4-benzothiazines of structure l 2we observed that their treatment with NaBH4, in neat acetic acid as solvent, gave rise to the expected dihydro-1,4-benzothiazines 2 or to the corresponding N-ethyl derivatives 3 (R = CH3) (Scheme I) depending on the amount of NaBH4 added. The unexpected formation of N-alkyl derivatives prompted us to extend the reaction to a number of primary and secondary amines, both aliphatic and aromatic, and to

3454

J. Org. Chem., Val. 40, No. 23,1975

Notes

Scheme I

Table I Results Obtained According to Procedure A Reactants Substrate

Carboxylic acid

la

Acetic Acetic Acetic Acetic Acetic Formic Acetic Acetic Acetic Propionic Acetic Propionic Acetic Acetic

lb

la IC

2a

R, - N - R 2 R,-NH

+

R2

RCOOH

+

Na0H4

-___ --_

CH2-R

4C

-'-.

4

5

R1- N - R 2

6

CO-R

a)

R,

b) Rl c)

R1

=C

~

H

= C6 H5 = R 2 = C6 H5

~;

R ~ =H

;

R 2 = CHg

d)

R1 I R 2 = -CHZ[CH211CH2-

e)

R,=

R2=

2c 4c 4d 4e 4a 4d 7a 7b

7

+

RCOOH

+

HaOHq

-

g-C4H9

8

several liquid carboxylic acids; invariably the corresponding N-alkyl derivatives were obtained (Table I). We noticed that by adding the amine to the NaBH4-carboxylic acid mixture previously refluxed for 3 hr, further heating for a period of 3 hr leads to the formation of N-acyl derivatives 6; addition of an alcohol or a phenol (7), in the place of the amine, yields, under the same conditions, the corresponding esters 8. In separate experiments, the volume of HZ1t3evolved on addition of NaBH4 to neat acetic acid was measured. The results of a number of experiments indicate that 3 mol of HPper mole of NaBH4 is immediately evolved a t 20', while approximately a fourth mole of H P i s evolved very slowly a t 20°, and more rapidly on heating at 80'. The remarkable ease and usefulness of this direct method of alkylating amines by means of the corresponding carboxylic acid has induced us to study the possibility of performing N-alkylations even with solid carboxylic acids dissolved in suitable solvents. In fact, it is clear from what has been reported so far that the method appeared limited to the use of liquid carboxylic acids. I t was first checked that the stoichiometry of the evolution of hydrogen was not changed when NaBH4 and carboxylic acid in the molar ratio 1:5 were allowed t o react in solvents such as benzene or toluene; in fact, 3 mol of hydrogen per mole of NaBH4 was evolved a t 20°, and approximately a fourth mole on heating under reflux. A certain number of reactions have therefore been performed in benzene yielding the N-alkyl derivatives 5 (Table 11). The addition of NaBH4 to solutions of the acids in benzene causes, in general, the formation of insoluble compounds very sensitive to moisture. We have isolated those from formic, acetic, monochloroacetic, and benzoic acid; these are high-melting solids (except that obtained from monochloroacetic acid, which decomposes a t 120-125');

2a 2b 3a 3C 3a 3C 5C 5d 5e 6a

BC 6d 8a 8b

(R (R (R (R (R (R (R (R (R (R (R (R

= = = = =

= = = = = = =

60 95 95 75 95 70 95 60 80 95 40 60 50 95

CH,) CH,) CH,) H) CH,) CH,) CH,) C,H,) CH,) C,H,) CH,) CH,)

Reactants Carboxylic acid

Reaction products

Yield, %

5a (R = CH,) 5a (R = C,H,) R = CZH, 4a Propionic 5 R, = C6H, RZ = n-C3H7 4b Propionic 5b (R = C,H,) 4b Palmitic 5b (R = CI,H,,) 4b Benzoic 5b (R = C6H,) 4c Acetic 5C (R = CH,) 4c Propionic 5C (R = C,H,) 4c Palmitic 5C (R = C,,H,i) 4c Monochloroacetic 5c (R = CH,Cl) 4c Monochloroacetic 5c (R = CH,Cl) a Reaction carried out with a molar ratio BHd-:substrate Molar ratio BHd-mbstrate 5 : l . 4a 4a

R1-0-CO-R

Yield, %

Table I1 Results Obtained According to Procedure B

Substrate

Rl-OH

Reaction products

Acetic Propionic

1

30 60 90" 65 70 20 70 35 25 35 90b of 10:l.

Table I11 Results Obtained According to Procedure C Reactants Substrate

Carboxylic acid

Reaction products

Yield, %

4b Benzoic 5b (R = C,H,) 90' 4c Acetic 5c (R = CH,) 80 4c Monochloroacetic 5c (R = CH,Cl) 100 N,N'-Diethyl 90b 2,2'-DithioAcetic d ianiline derivative 2,2'-DithioBenzoic iV,N'-Dibenzyl 70b dianiline derivative a Reaction carried out in toluene under reflux for 15 hr. Reaction carried out with a molar ratio BH4-:substrate of 1O:l. their ir spectrum exhibits a B-H band between 2480 and 2530 cm-l and two bands in the carbonyl zone. By treatment with water, they rapidly evolve Hz, and yield the carboxylic acid and boric acid in the molar ratio of 2:1, as shown by volumetric5 and potentiometric6 titrations; a third mole of carboxylic acid can be freed by addition of strong acids to the hydrolysis reaction mixture. These products are capable of carrying out N-alkylation reactions, when reacted with amines in solvents such as toluene or benzene (Table 111).

J. Org. Chem., Vol. 40,No. 23,1975 3455

Notes While the results so far available can hardly be regarded as conclusive, we feel that the experimental data support the assignment of the triacyloxymonohydroborate structure Na[(RC00)3BH] (9) to the N-alkylating species we have isolated. This structure agrees with the results of the hydrolysis: in fact, i t is obvious that on treatment with water 1 mol of 9 necessarily gives 1 mol of Hz,1 mol of boric acid, 2 mol of carboxylic acid, and 1 mol of the sodium salt of the carboxylic acid and therefore t h a t the addition of a strong acid is required to displace the third mole of carboxylic acid from the sodium salt. Furthermore, the structure 9 fits the ir spectroscopic data. In particular, the observed high frequency of B-H stretching bands seems appropriate for borohydrides, that can be looked a t as adducts of poor Lewis acids like the acyloxyboranes of the present case. This interpretation is consistent with the results in the paper by Rice et al.,7 correlating the increase in B-H stretching frequency of borane adducts with their lowered stability. Species 9 can form an aldehyde by intra- or intermolecular hydride reduction; indeed, we have obtained benzyl alcohol on refluxing 9 (R = C6Hb) for 20 hr in toluene and pmonochloroethyl monochloroacetate on refluxing 9 (R = CHZC1) for 6 hr in benzene. Furthermore, we found t h a t all species isolated from NaBH4 and carboxylic acids are capable of reducing iminium cations. These results seem to support the reaction pathway suggested by Gribble et a1.l Concerning the nature of acylating species, the evolution of 4 mol of Hz per mole of NaBH4 can be reasonably ascribed to the formation of a tetracyloxyborate Na[B(OCOR)4], 10. However, under the experimental conditions used, it is possible that species such as 10 decompose into RCOONa and B(OCOR)3, whose acylating ability has been reported.1,8

Experimental Sectiong N-Alkylation Reactions in Neat Carboxylic Acids. Procedure A. NaBH4 (0.1 mol) was added portionwise to neat carboxylic acid (30 ml), the temperature being kept at 20'; when the lively evolution of Hz had ceased, the amine (0.02 mol) was added. The reaction mixture was heated at 80' for 3 hr, cooled at room temperature, made alkaline with 2 N NaOH, and extracted with CHzClZ. The organic layer was separated, dried (Na2S04), and evaporated; PLC of the residue [light petroleum ether-ethyl acetate (9:l) as solvent] gave the corresponding N-alkyl derivative. The results obtained are reported in Table I. 4-Ethyl-2,2-dimethyl-3-phenyl-3,4-dihydro-2~-benzo[b][l,l]thiazine (3a, R = CH3): mp 103-104' (2-propanol); NMR (CDCl3) 6 7.6-6.6 (group of signals, 9 H, aromatic H), 4.27 (s, 1 H, C3 H), 3.8-3.0 (m, 2 H, CH2), 1.52 (8, 3 H, CZCHs), 1.12 (s, 3 H, C2 CH3), and 1.12 ppm (t, 3 H, CH3). 1l-Ethyl-5a,6,7,8,9,1O,lOa,ll-octahydrobenzo[ blcyclohepta[ e][ 1,llthiazine (3c, R = CH3): bp 120-122' (0.05 Torr); NMR (cc14)6 7.3-6.4 (group of signals, 4 H, aromatic H), 3.7-3.0 (group of signals, 4 H, NCHz + Csa H + Cloa H), and 2.5-0.8 ppm [group of signals, 13 H, CH3 + (CHZ)~]. 11-Methyl-5a,6,7,8,9,10,lOa,ll-octahydrobenzo[ blcyclohepta[ e][ Lllthiazine (3c, R = H): mp 56-57' (EtOH); NMR (CC14) 6 7.2-6.5 (group of signals, 4 H, aromatic H), 3.8-3.5 (m, 1, C1oa H), 3.5-3.3 (m, 1 H, C k a H), 2.87 (s, 3 H, CHa), and 2.4-1.2 ppm [group of signals, 10 H, (CHZ)~]. In the case of the imines 1, if the above procedure (A) is modified by allowing NaBH4 and benzothiazine to react in the molar ratio 1:l and reducing the heating time to 1 hr, the reaction products are the dihydro derivatives 2.2 The reaction of NaBH4 with neat carboxylic acids at 20' was explored in order to determine the amount of H2 evolved. In several experiments, it was found that 3.0 f 0.1 mol of H2 per mole of NaBH4 was formed. In a typical run, NaBH4 (0.079 g, 2.1 mmol) was added to neat acetic acid (0.8 ml) at 20'; 6.4 mmol of Hz was immediately evolved. Acylation Reactions in Neat Carboxylic Acids. NaBH4 (0.1 mol) was added portionwise to neat carboxylic acid (30 ml), the re-

action mixture was refluxed for 3 hr, the amine, the alcohol, or the phenol (0.02 mol) was added, and the reflux was prolonged for an additional 3 hr. By working up as above, the corresponding amide or ester was isolated. The results of these reactions are reported in Table I. The reaction of NaBH4 with liquid carboxylic acids under reflux was explored in order to determine the amount of HZevolved. In several experiments, it was found that 3.8 f 0.1 mol of Hz per mole of NaBH4 was formed. In a typical run, NaBH4 (0.125 g, 3.3 mmol) was added to neat acetic acid (1.5 ml); after reflux for 3 hr, 12.7 mmol of H2 was evolved. N-Alkylation Reactions in Benzene Solution. Procedure B. NaBH4 (0.1 mol) was added to a solution of the carboxylic acid (0.33 mol) in dry benzene (50 ml), the temperature being kept at 20°. When the evolution of Hz had ceased, the amine (0.05 mol) was added, and the reaction mixture was refluxed for 3 hr, cooled at room temperature, and shaken with 2 N NaOH. The organic layer was separated, dried (NazSOd), and evaporated; PLC of the residue [light petroleum ether-ethyl acetate (9:l) as solvent] gave the corresponding N-alkyl derivative. The results obtained are reported in Table 11. It should be noted that the reported yields were not optimized; in fact, better yields can be attained by raising the BHd-:substrate molar ratio and/or the refluxing time, as shown in the case of 5c (R = CHZCl), which was obtained in 90% yield using a molar ratio BHr-:substrate of 51. N-(2-Chloroethyl)-N-phenylaniline(5c, R = CHzCl): bp 110-112' (0.1 Torr); NMR (CCld) 6 7.2-6.9 (group of signals, 10 H, aromatic H) and 4.3-3.5 ppm (group of signals, 4 H, -CHzCHz-). N-Hexadecyl-N-phenylaniline(512,R = CL5H31): mp 43-44' (EtOH); NMR (CC14) 6 7.6-6.9 (group of signals, 10 H, aromatic H), 3.78 (t, 2 H, NCHz), and 2.0-0.6 ppm (group of signals, 31 H, aliphatic H). The reaction of NaBH4 with solutions of carboxylic acids in dry benzene or toluene at 20' was explored in order to determine the amount of Hz evolved. In several experiments, it was found that 3.0 f 0.1 mol of H2 per mole of NaBH4'was formed. In a typical run, NaBH4 (0.144 g, 3.8 mmol) was added to a solution of acetic acid (1.5 ml) in dry toluene (10 ml) at 20'; 11.1 mmol of H2 was evolved. Sodium Triacyloxymonohydroborates, 9. NaBH4 (0.1 mol) was added portionwise to a solution of the carboxylic acid (0.33 mol) in dry benzene (1l.), the temperature being kept at 20'. After hydrogen evolution, the precipitate 9 was rapidly collected by suction, washed with ethyl ether, and dried under vacuum. Compounds 9 do not melt up to 300°, except for 9 (R = CHzCl), which decomposes at 120-125'. 9 (R = CHZCl): ir 2530 (B-H) and 1735,1685 cm-l (C=O). 9 (R = H): ir 2480 (B-H) and 1680 cm-I (broad band, C=O). 9 (R = CH3): ir 2480 (B-H) and 1660 cm-I (broad band, C=O). 9 (R = C&,): ir 2490 (B-H) and 1670,1635 cm-' (C=O). A suspension of 9 (R = C6H5) in dry toluene (50 ml) was refluxed for 20 hr and cooled at room temperature, and the solid was filtered off. The presence of benzyl alcohol in the filtrate was monitored by vapor phase chromatography. Similarly, @-monochloroethyl monochloroacetate was formed by refluxing a suspension of 9 (R = CHZCl) in dry benzene for 6 hr, and monitored by vapor phase chromatography. Compounds 9 are capable of reducing imonium cations. In a typical run, 9 (R = CH3) (0.22 g) was added to a stirred suspension of la HCl (0.29 g) in dry benzene (50 ml) at room temperature. ACter 10 min, the reaction mixture was shaken with 2 N NaOH, and the organic layer was separated, dried (NaZSOd), and evaporated to give 2a (0.25 9). Hydrolysis of the Triacyloxymonohydroborates 9. On reaction with an excess of water, 9 promptly release hydrogen; in a typical run, 0.400 g of 9 (R = C6H5) (1.00 mmol) evolves 0.98 mmol of Hz. The hydrolysis mixtures were examined by potentiometric titration with 0.1 N NaOH; a molar ratio of carboxylic acid to boric acid of 1.95 f 0.051 was determined. In the case of benzoic acid derivative (9, R = C&,) a gravimetric determination was also accomplished. 9 (R = C&) (3.91 g) was hydrolyzed in hot water (150 ml); after cooling, the reaction mixture was extracted with CCL, and the organic layer was separated, dried (NazSOd),and evaporated; a residue of benzoic acid (2.26 g) was obtained. The aqueous layer was strongly acidified with HC1, and extracted with CCL; the organic phase was separated, dried (NazS041, and evaporated; a residue of benzoic acid (1.25 g) was obtained. N-Alkylation Reactions Carried Out with 9. Procedure C.

3456 J. Org. Chem., Vol. 40, No. 23,1975 T h e amine a n d 9 in t h e m o l a r r a t i o 1:5 were refluxed in dry benzene for 6 hr. By w o r k i n g up as for t h e above procedure (B) t h e corresponding N - a l k y l derivatives were isolated. T h e results obtained are reported in T a b l e 111.

Acknowledgment. The authors are grateful to Professor

V. Carelli for his helpful suggestions. This work was sup-

Notes males.2 As part of a program of research on the properties and functions of cockroach pheromone^,^ we undertook and now describe a synthesis of 1. After completion and submission of this work for publication, an account with limited experimental details of a synthesis of 1 along somewhat similar lines by Ishii and coworkers appeared?

ported by a research grant from CNR, Rome, Italy. Registry No.-la, 49634-65-3;lb, 19195-32-5;IC, 25069-59-4; 2a, 49634-66-4;2b, 25069-68-5;2c, 25069-64-1;3a ( R = CHJ), 56553-67-4; 3c ( R = H), 49634-64-2; 3c ( R = CHs), 56553-68-5; 4a, 62-53-3; 4b, 100-61-8; 4c, 122-39-4; 4d, 110-89-4; 4e, 111-92-2; 5 (R = Et; R1 = Ph; Rz = Pr), 2217-07-4; 5a ( R = CHd, 103-69-5; 5a (R = CzHj), 622-80-0;5b ( R = CzH5), 13395-54-5;5b ( R = CsH5), 614-30-2; 5b ( R = C15H31), 56553-69-6; 5c (R = CHs), 606-99-5; 5c (R = CHZCl), 42393-65-7;5~ ( R = C1,5H31), 51580-76-8;5d (R = CH3), 766-09-6; 5e ( R = CH3), 4458-33-7; 6a ( R = CzHs), 620-71-3; 6c ( R = CH3), 519-87-9; 6d (R = CzHj), 14045-28-4; 7a, 135-19-3; 7b, 91-01-0; 8a ( R = CH3), 1523-11-1; 8b ( R = CH3), 954-67-6; 9 (R = H), 56553-59-4;9 ( R = CHa), 56553-60-7;9 ( R = CH&l), 56553-61-8; 9 ( R = C & , ) , 56553-62-9; acetic acid, 64-19-7; formic propionic acid, 79-09-4; p a l m i t i c acid, 57-10-3; benacid, 64-18-6; monochloroacetic acid, 79-11-8;2,2’-dithiodianizoic acid, 65-85-0; line, 1141-88-4; N,N’-diethyl-2,2’-dithiodianiline, 56553-70-9;

N,N’-dibenzyl-2,2’-dithiodianiline, 56553-71-0.

R e f e r e n c e s a n d Notes (1) G. W. Gribble, P. D. Lord, J. Skotnicki, S. E. Dietz, J. T. Eaton, and J. L. Johnson, J. Am. Chem. Soc., 96,7812 (1974). (2) V. Carelli, P. Marchini, M. Cardellini, F. M. Moracci, G. Liso, and M. G. Lucareiii, Tetrahedron Left., 4619 (1969); Ann. Chim. (Rome), 59, 1050 (1969); Int. J. Sulfur Chem., 6, 267 (1973). (3) It has been reported4 that reaction of NaBH4 (1 mol) and carboxylic acids (1 mol) leads to evolution of HP(1 mol) and to the formation of monoacyloxytrihydroborate Na [(RCOOBH3)]. (4) H. C. Brown and B. C. Subba Rao, J. Am. Chem. Soc..82, 681 (1960); T. Reetz, hid., 82, 5039 (1960). (5) W. Gerrard, M. F. Lappert, and R. Shafferman, J. Chem. Soc.,3648 (1958). (6) J. Mohay and J. Mohay-Farkes, Acta Pharm. Hung., 37, 71 (1967); Chem. Abstr., 66, 9 8 5 5 6 ~(1967). (7) B. Rice, R. J. Galiano, and W. J. Lehmann, J. Phys. Chem.. 61, 1222 (1957). (8) A. Pelter and T. E. Levitt, Tetrahedron, 26, 1899 (1970). (9) Ail melting points (determined on a Kofler apparatus) and boiling points are uncorrected. The compounds previously described were compared with authentic samples. All new compounds, but 9, gave satisfactory elemental analyses. ir spectra as Nujol mulls were recorded on a PerkinElmer Model 257 grating spectrophotometer. NMR spectra were obtained on a Jeol 60 spectrometer, using Me4Si (6 0 ppm) as internal standard. The reaction products were Isolated by preparative thin layer chromatography (PLC) on Merck PF254 silica gel coated plates. Commercial NaBH4 was recrystallized according to Brown et at.’’ (IO) H. C. Brown, E. J. Mead, and 8. C. Subba Rao, J. Am. Chem. Soc., 77, 6209 (1955).

CH3

CH3

I

I

nC,sH37CH(CH2)761HCOCH3 1

As starting material in our synthesis we employed 8-oxononanoic acid (2)) prepared from e-caprolactone by modifications of the route of Kameoka et al.5 via 6-bromohexanoic acid, esterification, and acetoacetic ester synthesis. Treatment of the methyl ester 3 in polar solvent with the Wittig reagent derived from octadecyltriphenylphosphonium bromide gave, in 48% yield, methyl 8-methyl-8-hexacosenoate (4))apparently mainly (by GLC) the 2 isomer.6 Hydrogenation of 4 afforded the saturated ester 5, which, after LiAlH4 reduction to the corresponding alcohol 6 and conversion into the bromide 7, was used to alkylate diethyl methylmalonate. Hydrolysis of the resulting diester 8 and decarboxylation of the acid 9 gave 2,10-dimethyloctacosanoic acid (lo),which, with two widely separated asymmetric centers, was undoubtedly a mixture of the two possible diastereoisomers. Treatment of 10 with 2 mol of methyllithium then furnished the desired ketone 1, mp 28-31O (lit.4 29-31’), in 50% overall yield from the Wittig product 4. CH3

I

CH,CqCH2)&02R 2,R-H 3,R = CH,

nC17H&H=C(CH&CO&!HS 4

CH3

I

I R2 8, R, = Rz = COzC2H, 9, R, = Rz= COBH 10, R1 = C02H;& = H

5, R = C02CH, 6, R = CHzOH 7, R = CHzBr

CH3

I

nG18H37dHCOCH3 11

Synthesis of 3,l l-Dimethyl-2-nonacosanone, a C o n t a c t C o u r t i n g P h e r o m o n e of t h e G e r m a n Cockroach’ A l b e r t W. Burgstahler* a n d L e l a n d 0. Weigel

Department of Chemistry, The University of Kansas, Lawrence, Kansas 66045 William

J. B e l l a n d M i c h a e l K.R u s t

Department of Entomology, The University of Kansas, Lawrence, Kansas 66045 Received April 29,1975

From the cuticle of sexually mature female German cockroaches ( B l a t t e l l a g e r m a n i c a ) , Ishii and coworkers recently isolated a contact chemoreceptive agent, identified as 3,11-dimethyl-2-nonacosanone ( I ) , which was shown to elicit typical courting behavior, including wing raising, in

Although spectral data indicate that the synthetic and natural ketones are structurally identical, our method of synthesis, like that of Ishii and coworkers,4 undoubtedly afforded a mixture of the two possible diastereoisomers of 1. It is not surprising, therefore, that the natural product has a different melting point (45-46’), even though it appears to be optically i n a ~ t i v e . ~ . ~ Previous studies have shown that courting behavior in the German cockroach includes antennation (antennal stroking) of the female by the male, presumably allowing the latter to perceive sex pheromone on the cuticular surface of the female.7 In our bioassay of synthetic 1 we used antennae ablated from American cockroaches ( P e r i p l a n e t a a m e r i c a n a ) to eliminate any possible stimuli associated with German cockroach antennae. Control antennae, dipped only in carbon tetrachloride, evoked no response. Antennae dipped in a 70 wg/ml solution of synthetic 1 in carbon tetrachloride elicited typical wing raising and other features of courting display7 in 5%of a group of males ( n = 60) kept isolated from females. A t a higher concentration of 500 pg/ml the response was 70% in a group of isolated